Robust Self-Localization in Industrial Environments based on 3D Ceiling Structures

Author

Wulf, Oliver ; Lecking, Daniel ; Wagner, Bernardo

Author_Institution

Inst. for Syst. Eng., Hannover Univ.

fYear

2006

fDate

9-15 Oct. 2006

Firstpage

1530

Lastpage

1534

Abstract

This paper introduces a new perceptual model for Monte Carlo localization (MCL). In our approach a 3D laser scanner is used to observe the ceiling. The MCL matches ceiling structures like beams, columns, air condition and lightning installation against a world model containing line and point features. Thus the localization is not effected by clutter or any kind of dynamic obstacles on the ground level. Different experiments show that our system can be used to localize robustly in factory buildings and halls. Experimental results include a 5 day run of an autonomous fork lift truck on a logistics fair in Hannover

Keywords

Monte Carlo methods; ceilings; industrial robots; mobile robots; optical scanners; path planning; 3D ceiling structures; 3D laser scanner; Monte Carlo localization; autonomous mobile robots; factory buildings; industrial environments; robust self-localization; Cameras; Laser modes; Laser noise; Mobile robots; Monte Carlo methods; Production facilities; Robot kinematics; Robustness; Servomechanisms; Vehicle dynamics; 2D feature map; 3D laser scan; Monte Carlo Localization; dynamic environments;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Robots and Systems, 2006 IEEE/RSJ International Conference on

Conference_Location

Beijing

Print_ISBN

1-4244-0258-1

Electronic_ISBN

1-4244-0259-X

Type

conf

DOI

10.1109/IROS.2006.281984

Filename

4058590